[s-cars] RE: [Audi20V_Kruemmer] Re: tubular exhaust manifold and turbo options

[MLP]

Minhea, what volumetric efficiency assumption(s) do you make for the AAN
motor?

At the Mitsu 3000GT site (www.stealth316.com - Jeff Lucius et al.)
appear to have worked VE efficiency backwards off a torque curve
published by Mitsubishi for the 3000GT motor.  Using that info, Mssrs.
Lucius & Joe Gonsowski et al, estimated a stock 3.0 liter V6 motor's
efficiency at varrious rpms as: 

      Stock  Modified 3000GT
rpm    VE       VE
2K     91%     ? NG
3K     95%     88%
4k     93%     93%
5k     90%     96%
6k     81%     94%
7k     75%     90%
8k     NG      85%

The other compressor map quirk of note, is the steal316.com sites
exclusive uses of & universal adaptation of the various maps presented
to cfm for plotting engine volumetric efficiency curves against
compressor maps.  For the Garrett maps, it's based on 10 lbs/min = @ 145
cfm.  Lucius description for converting the Garrett requires knowing the
temp & presure at which the lbs/min were charted.   

In most Garrett maps the conversion factor (based on 85.31°F & 28.4" Hg
or 13.9487 psi) is 10 lb/min ='s 144.7178 cfm.  If this is the
"necessary & proper" way of sizing a turbo, i.e. converting all flow
figures to cfm, essentially recalibrating the maps x-asis to conform to
VE estimates, it certainly makes using the Garrett, and any other
manufacturer, i.e. KKK, compressor maps a royal pain to use.

FWIW, if one looks at the KKK compressor maps available on sjm's site,
they appear to use two (2) scales marked off along the x-axis, and like
the Garrett, a temperature (293°Kelvin), and pressure, 981 mbar for
making the same conversion from mass, probably Kg/sec or minute, to
m^3/sec.  If in fact that's the case (i.e. the upper x axis scale is in
m^3/sec, the lower in Kg/sec), that would make the KKK maps far easier
to use.  

Turbonetics provides a pretty ineffective pair of "charts" (one for 5 to
10 psi, the second 10 to 15 psi) in its catalog for "converting"
displacement flow (i.e. Assume VE of 100%m 70% compressor efficiency,
80°F, sea-level and an 80% efficient IC,.... your air flow demand in
lbs/min will be on the Y axis, for a given displacement of xxx as picked
of the x-axis)  

Unless I'm missing something, Innovative doesn't bother with the lbs/min
to cfm conversion discussion.  Their catalog just cuts to the chase
suggesting that what one needs to do is decide how much horsepower is
wanted based on a rough rule of thumb, 1 lb of air at 100% VE ='s 10 hp,
for 300bhp one is going to need about 30lbs of air.  Not much is said
about avoiding the dashed line over to the left on the maps, or even why
one would be looking to the left, when well 30+ lbs is well over to the
right of the dashed line :-).

Finally for those who get this far, the Tuesday AM "If you didn't get
enough caffine, maybe this will wake you up" site or suggestion of the
day, http://www.grc.nasa.gov/WWW/Oilfree/turbocharger.htm :-)

mike

-----Original Message-----
From: Mihnea Cotet [mailto:mik at info.fundp.ac.be] 

Yeah, I hadn't figured either how great a wheel the gt40 is!

Your post only shows you have no clue what a compressor map is, nor how
it 
works.

Try to plot 2.6PR at 3500 RPM, 2.9 at 4000, 4500 and 5000 RPM for a 2.2 
liter engine and see how much air flows at that pressure.

This is getting real tricky when people have a direct financial interest

into reselling Garrett turbos, which aren't bad turbos by all means, but

rather it's hard to know what to buy when even a vendor recommends
people 
something that will never work on our engines.

No offense intended,

Mihnea